Skip to main navigation Skip to search Skip to main content

Can LLMs Detect Display Issues? Uncovering the Impact of Prompting Techniques

    • Jeonbuk National University

    Research output: Contribution to conferenceConference paperpeer-review

    Abstract

    While Large Language Models (LLMs) offer a promising, cost-effective alternative to conventional methods for detecting UI display issues, the effectiveness of various prompting techniques has not been systematically analyzed. This paper investigates the impact of various prompting techniques on the performance of LLMs in identifying UI display issues. Our analysis reveals that while some techniques significantly boost detection performance, others show minimal impact, and certain issue types, like Misalignment, remain challenging for all tested approaches.

    Original languageEnglish
    Title of host publicationUIST Adjunct 2025 - Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology
    EditorsAndrea Bianchi, Elena Glassman, Shengdong Zhao, Jeeeun Kim, Ian Oakley, Wendy E. Mackay
    PublisherAssociation for Computing Machinery, Inc
    ISBN (Electronic)9798400720369
    DOIs
    StatePublished - 2025.09.27
    Event38th Annual ACM Symposium on User Interface Software and Technology, UIST 2025 - Busan, Korea, Republic of
    Duration: 2025.09.282025.10.1

    Publication series

    NameUIST Adjunct 2025 - Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology

    Conference

    Conference38th Annual ACM Symposium on User Interface Software and Technology, UIST 2025
    Country/TerritoryKorea, Republic of
    CityBusan
    Period25.09.2825.10.1

    Keywords

    • Display Issue
    • Large Language Model
    • Prompt Engineering
    • UI

    Fingerprint

    Dive into the research topics of 'Can LLMs Detect Display Issues? Uncovering the Impact of Prompting Techniques'. Together they form a unique fingerprint.

    Cite this